New Revised model

The revised model will continue to strengthen Davis’s original TAM model as the only two independent variables that has a significant contribution to citizens intention to use. However, this does not mean that the other variables are useless but may rather play a role as

antecedents to PEOU and PU. By looking at figure 8 one can see a visualisation on how these variables might play a role and how they influence each other.

One can logically explain why Awareness, Risk, Trust in Technology and Trust in

Government Technology had no significant and direct influence on citizens intention to use, which was the case in the original model. As previously mentioned, without awareness of a service available, it will be difficult for users to see the need of it. Awareness is people’s knowledge of technology and the availability of electronic services (Venkatesh et al., 2003).

Since the mobile application referred to in this study is a service that has not been experienced and that people are not fully aware of it will be difficult to have a direct link to their intention to use the service. And without the awareness of a service it will be difficult for the end-user to have an opinion if they have the necessary facilities, resources, equipment and assistance to support the use of the mobile application. This is also the case for its compatibility, without awareness about the service, it is difficult to say if the technology is compatible to the end-user. Hence, awareness has an impact on Facilitating Conditions and Compatibility.

Perceived risk is defined as a consumer’s perceptions of the uncertainty and the possible undesirable consequences of buying a product or service (Fagih, 2011). In this model one sees that perceived risk has a direct impact on end-user’s trust in technology and their social influence. The behaviour of citizens is heavily affected by perception of risk. End-users are frequently uncertain as to the implications or consequences of a decision or action

(Almuraqab and Jasimuddin, 2017). Referring to the Corona example from the theory chapter one could see that end-users didn’t want to share their personal data due to the risk and trust in the given technology. Referring to appendix 6 one can see that with higher risk involved it has a significant impact on their social influence.

Trust in technology and trust in government technology has a significant impact on each other and is not surprising since these factors measures a correlated concept.

With rationale thinking one can argue that without trust in technology you will most likely not have trust in the technology provided by the government or local municipality. Easily put, you need trust in technology in order to have trust in government technology. This is also

statistically proved in this study. A government that faces greater mistrust and suspicion may discover its citizens find ways to ignore and resist its actions and are suspicious of its

pronouncements and policies. Previously research has also found a higher level of trust in government correlate with more intensive e-government service use, indicating that those who are satisfied with such services are also more trusting of government. Hence, citizens are less reluctant to provide information if they are confident that it will not be misused by authorities, obtained by private parties through security failure, corruption or used in ways not intended by, and against the interests of them (Horsburgh et al., 2011). Trust in technology and government technology is therefore an important factor for governments and local municipalities when providing services and new technology to its citizens. This was also tested in the revised model by incorporating trust in technology and trust in government technology to the TAM variables. However, based on tests from the new model trust in government technology only had an influence on PU and not PEOU, and can therefore play a role as an antecedent to PU. Again, the example with the mobile application Smittestopp gives an indication on how reluctant citizens can be in sharing their personal information and trustworthiness into the given technology, and also shows that both risk, trust in technology and trust in government technology have an impact on the TAM framework on both attitude and intention to use. This also supports Belanche’s (2012) findings of integrating Trust in the TAM framework.

The other antecedents are not logically connected to the TAM variables but have been tested and proved to have an impact based on previously research. According to Venkatesh et al (2008) facilitating conditions have limitations in capturing the effects of external impediments to system use. Facilitating conditions does not directly affect the use of the system, because this variable reflects recognition of the existence of (or lack of) favourable conditions. They argue that the use of the system depends on whether and to what degree an individual expects that facilitating conditions will enable system use, considering other potential behavioural impediments. The measure of facilitating conditions in this study mainly focuses on support

and assistance for users, however, it does not take into account individual’s beliefs about the ease of use and usefulness of a given technology. Thus, in the new model facilitating

conditions are evaluated in the light of technological impediments which can be captured both through perceived ease of use and perceived usefulness. However, since FC had no significant impact on PU, it was removed from the model.

Venkatesh et. Al (2003) define social influence as the degree to which an individual perceives that “important others believe” (for example friends and family) they should use the

technology and that social influence is crucial in shaping user behaviour. The paper of Sathye et al (2018) found SI to have a significant positive association with intention to use. However, they did not only demonstrate that SI had a positively impact on intention to use, but that its impact gets transmitted through the constructs of PU and PEOU. This is also the case in the new revised model as it indicates that social networks (family and friends) has a positively and significantly impact on PU and PEOU. Indicating that their findings support the findings of the new model. Social influences significantly impact individual perceptions about

usefulness and perceived ease of use of technology (Sathye et al., 2018).

Since Perceived compatibility had no significant impact on citizen intention to use in the original model, I have argued that it may be seen as an antecedent to the strongly supported TAM variables. From my findings one can see that Compatibility has a significant impact on PU and not far from a significant impact on PEOU (.017) with a 95 % confidence interval, see appendix 6.

These findings are supported by the research of Isaac et al. Their findings validated perceived compatibility as an antecedent variable on TAM (2016). Indicating that the more the citizens found the mobile application and technology to be consistent and compatible with their beliefs, values, lifestyle and needs, the more they would see the application as easy to use, flexible, understandable and can be used to accomplish their tasks quicker and easier.

Figure 9: New Revised Model

(Chi-Square X² = 510.655, df = 332, p = .000, X²/df = 1.538, RMSEA = 0.67, 90 % Confidence Interval: .055, .078, CFI = 913)

From the new and revised model one can see that there is support for the TAM variables PU and PEOU on citizens intention to use digital communication technology. However, PEOU

has a P level of .072 on intention to use which means that it is not significant with a 95 % confidence level. Nonetheless, it does not mean that one rejects it’s influence on intention to use but rather focus on it as a measure of precision. According to Hardy and Bryman “as when setting confidence intervals, there is nothing sacrosanct or magical about these numbers, either Z or alpha. They are entirely conventional choices, and one is free to select a different number. Typically, one begins with a value of alpha that is personally acceptable and that will be acceptable to one's audience” (2009). “Increasing the significance level to a higher value (e.g. .10) allows for a larger chance for being wrong, but also makes it easier to conclude that the coefficient is different from zero” (Joseph F. Hair et al., 2009)

In this new revised model, a confidence interval of 90 % is set. Keeping in mind that the study measures citizens intention to use a service, and not a health study for example where an Alpha above 0.01 (99%) would not be accepted.